1. Field of the Invention
This invention relates generally to wireless data communications, and more specifically to a method of substantially improving error correction capability of block codes in desired data communication applications.
2. Description of the Prior Art
In Third Generation Partnership Project (3 GPP), the transport format combination indicator (TFCI) is used for properly decoding the received data. The TFCI indicates one of the choices, namely, one of the transport format combinations regarding how to decode the channel bits. The Node-B (base station) informs the mobile station of the set of available transport format combinations. The TFCI determines coding type, CRC bits, transport block size, etc. Without the proper TFCI, the decoded data will most certainly fail CRC and the transmission will be counted as a block error. The TFCI field is encoded by using a Reed Muller code to protect it.
FIG. 1 shows the frame structure 10 for a downlink dedicated physical channel (DPCH). Depending on the slot format, there may be anywhere from 0 to 8 TFCI bits per slot in normal mode, or 0 to 16 bits per slot in compressed mode. This results in 0 to 120 bits for TFCI in a frame. Whenever the number of TFCI bits after encoding exceed or are less than the number of available TFCI bits in a frame, the TFCI bits are punctured or repeated, respectively. In the case of normal or logical split TFCI mode, the TFCI is encoded using a (32,10) Reed Muller code. The encoding procedure 20 is shown below in FIG. 2. In case of split TFCI using hard split, the TFCI for the DPCH and Downlink Shared Channel (DSCH) channels are encoded separately using a (16,5) bi-orthogonal code. FIG. 2 shows the encoder block diagram for the Normal mode.
The reception of the TFCI is critical to proper operation of the user equipment (UE). Whenever the TFC is received with an error, data for the corresponding transmission time interval (TTI) are most likely to be discarded. For 80 ms TTI, that is the data that has been received over the last 8 frames and translates into large chunks of data being lost. One current universal mobile telecom system (UMTS) chipset uses the Green Machine disclosed by F. J. MacWilliams and N. J. A. Sloane, The theory of Error-Correcting Codes, North-Holland, Amsterdam, 1977, to decode the encoded data transmitted by the UTRAN. In this report, a technique is proposed to improve the performance of the TFC decoder using an alternate approach. Although this technique is used to decode Reed Muller encoded data, it can be easily extended to other codes whenever all the code words of a code are not used.
The Green Machine decoder is used to decode Reed-Muller (RM) encoded data as stated herein before. The green machine is an approach to decode RM codes using Walsh Hadamard Transforms. In this approach, the decoder searches for the decoded data among all the code words even if a large subset of these code words is not used. In addition, the basis vectors for the RM code need to be saved in memory along with intermediate summation results to decode the data properly. The block diagram of the green machine decoder structure 30 is shown in FIG. 3.
Another approach used in the prior art is the generator matrix approach that uses a hard decision approach to carry out syndrome testing such as disclosed by Proakis, John G., Digital communications, McGraw-Hill, Inc., 1995.
In U.S. Pat. No. 6,341,125, entitled Method for improving TFCI transportation performance, by Hong, Sung Kwon, et al., issued Jan. 22, 2002, an attempt is made to improve the encoder to transmit TFCI with a low BER. In any of the above, no attempt is made to improve the decoder's performance or to provide more protection to the transmitted bits.
In view of the foregoing, it would be both advantageous and desirable in the wireless data communication art to provide a method that is independent of the encoder and that can be used to decode data encoded by any block code in a manner that substantially improves the error correction capability of the block codes. It would be further advantageous and desirable if the structure associated with the method could allow the testing of those hypotheses that are known to exist, such that one can use the a priori knowledge of the possible set of hypotheses to only search from among them. Such a method would also be advantageous and desirable since knowing the subset of the code word space that is being utilized in essence allows the distance between the code words to be increased yielding significant decoding benefits.